Cofilin is a well-documented promoter of actin turnover in all eukaryotic cells. It undergoes dephosphorylation (activation) and oxidation to dimers when neurons are stressed by agents associated with Alzheimer disease (AD), all of which increase reactive oxygen species (ROS). This cofilin oxidation results within neurites in the formation of rod shaped cofilin-saturated actin bundles (rods). Rods immediately after their appearance are energy conserving, but when sustained causes neurite degeneration by blocking transport. Agents can induce rods via mitochondrial inhibition that generates ROS (e.g., excitotoxic glutamate) or via a prion-dependent pathway, probably involving membrane NADPH oxidase (NOX) activation. The latter pathway is activated by the proinflammatory cytokine TNF as well as by soluble SDS-stable dimers/trimers of Amyloid- (Ad/t), the A form that correlates best with dementia severity. Cofilin-actin rod formation causes the synapse elimination without neuronal cell death that is characteristic of amnestic mild cognitive impairment, an early stage in progression to Alzheimer disease. Our hypothesis is that it is cofilin-actin rods that initiate and exacerbate synaptic dysfunction typical of both sporadic (SAD) and familial AD (FAD). To demonstrate definitively that rod formation per se contributes to cognitive decline associated with AD, we need to develop a mouse model resistant to rod formation. Only a rod resistant mouse will allow us to answer the critical question: Do rods, themselves, cause synaptic loss or is synaptic loss due to stress-induced changes other than rod formation? It should be possible to make such a model since we have characterized a non-rod forming mutant of cofilin (K22Q), which is able to rescue normal behavior of cofilin-silenced cells as well as wild type cofilin. Three strategies are described to make a knock-in mouse in which cofilin K22Q will replace wild type cofilin. Two of these will make a conditional mouse which expresses wild type cofilin until mice are given tamoxifen. In these mice tamoxifen activates expression of Cre recombinase, which will initiate the inactivation of the wild type gene and the activation of the K22Q cofilin gene. The project is high risk since obtaining such a mouse is not guaranteed. Ultimately these mice will be used in behavioral assays to assess their cognitive ability under normal and stress conditions mimicking SAD or FAD. Finding that rods per se are necessary for synapse loss and cognitive impairment would make this project high reward because we already have identified a nutraceutical, the pentacyclic triterpene ursolic acid (UA), that blocks and reverses Ad/t- and TNF-induced rods in cultured neurons and reverses oxidative stress markers and cognitive deficits in a brain oxidative stress mouse model. Thus we will determine if UA functions by reducing cofilin pathology and if it can be a major therapy for reducing cognitive deficits in mouse models of both SAD and FAD.